That's pretty awesome query... have to try it out though...

Thanks a lot for your time!

Both scripts are OK, thank you very much 🙂

That works, thanks Pyay!

Thanks Pyay, almost what I want. With your query I get this:

1 10.00 10.00 2009-06-02 00:00:00.000

1 10.00 15.00 ...

Hey, of course this are sample data 🙂

Maybe this is better representation of actual data:

if object_id('tempdb..#temp') is not null drop table #temp

create table #temp (item varchar(10), price1 money, price2 money,...

I see I wrote incorrect result in my first post, now it is correct.

Christopher and Pyay, your query returns only distinct sets of item, price1 and price2. That's not ok...

With 'ordered by date desc' I mean that if there are two rows with same data (item, price1 and price2), then the returned row should be the one with greater...

The only reason I see why somebody would need this kind of comparison is solving some of the problems from Project Euler/[/url]. There are a lot of problems involving combinations...

Thanks, that explains a lot. It looks like I have to learn execution plans prior to indexes... now I am studying Execution plans on SQL in the Wild, nice and...

Maybe you can create a view based on your query and then select data from it?

OK, I removed unique constraint on Customer table.

Ramesh, you're right, I will change it back, it is more readable that's way 🙂

This is my own Project for learnign purposes (IS...

Thanks for your reply, I changed my code, hope it is OK now:

create table [Country](

[CountryID] int identity(1,1) not null,

[Country] varchar(50) not null,

...

I think you can use breakpoints in SQL Server 2008.

In 2005, no.

You can put 'print' commands in SP, run SP and watch for output (in Management Studio)

Or use xp_logevent in SP for writing to eventlog

Or insert data inside of SP into...

Try something like this:

select last_user_update

from sys.dm_db_index_usage_stats

where database_id = db_id( 'databasename' )

and object_id = object_id('tablename')